People
Dr Fabricio Goes
Lecturer in Creative Computing
School/Department: Computing and Mathematical Sciences, School of
Profile
My research interest in Large Language Models applied to Computational Creativity. Due to my recent publications, I was recently invited to be interviewed on the Data Skeptic Podcast, the most popular data science podcast, the CBC Spark podcast and the restofworld.org tech news to talk about the use of LLMs to evaluate creativity. I've also recently been invited to be a reviewer in Nature Human Behaviour. I'm also a HEA and IEEE member.
I'm looking for PhD students with interests in the application of LLMs in the creative industries. Please, contact me if you are interested.
Latest publications
- Is GPT-4 Good Enough to Evaluate Jokes? ICCC, 2023.
- Pushing GPT’s Creativity to Its Limits: Alternative Uses and Torrance Tests. ICCC, 2023.
- Bits of Grass: Does GPT Already Know How to Write like Whitman? ICCC, 2023.
- On the Power of Special-purpose GPT Models to Create and Evaluate New Poetry in Old Styles. ICCC, 2023.
- CreativeStone: a Creativity Booster for Hearthstone Card Decks. IEEE Trans. Games, 2023.
- Crowd score: A Method for the Evaluation of Jokes using Large Language Model AI Voters as Judges. ArXiv, 2022.
Research
I'm currently a member of the International Conference on Computational Creativity (ICCC) program committee, the most important conference in this hot topic. I supervise/co-supervise 3 PhD students on topics related to Computational Creativity and I'm currently looking for strong PhD candidates, please get in contact if you are interested.
[Selected papers on Computational Creativity]
Fabricio Goes, Piotr Sawicki, Marek Grzes, Dan Brown, and Marco Volpe. "Is GPT-4 Good Enough to Evaluate Jokes?", In International Conference on Computational Creativity (ICCC), 2023.
Fabricio Goes, Marco Volpe, Piotr Sawicki, Marek Grzes, and Jacob Watson. "Pushing GPT’s Creativity to Its Limits: Alternative Uses and Torrance Tests", In International Conference on Computational Creativity (ICCC), 2023.
Celso França, Zisen Zhou, Carolina Fernanda da Silva, Cibele Simões de Oliveira Santos, Lucas Braga Ferreira, Lucas Henrique Pereira, Marcos Pablo Souza de Almeida, Marina Iolanda Oliveira, Thaís Regina Damásio, Vinícius Silva Pacheco, and Fabricio Goes, "Creativestone: a Creativity Booster for Hearthstone Card Decks". IEEE Transactions on Games, 2023.
Piotr Sawicki, Marek Grzes, Fabricio Goes, Dan Brown, Max Peeperkorn, and Aisha Khatun. "Bits of Grass: Does GPT Already Know How to Write like Whitman?" In International Conference on Computational Creativity (ICCC), 2023.
Piotr Sawicki, Marek Grzes, Fabricio Goes, Dan Brown, Max Peeperkorn, Aisha Khatun, and Simona Paraskevopoulou. "On the Power of Special-purpose GPT Models to Create and Evaluate New Poetry in Old Styles", In International Conference on Computational Creativity (ICCC), 2023.
Fabricio Goes, Zisen Zhou, Piotr Sawicki, Marek Grzes, Daniel G. Brown, "Crowd Score: A Method for the Evaluation of Jokes using Large Language Model AI Voters as Judges", Arxiv, 2022.
W. Santos, J. Ribeiro, F. Magalhaes and Fabricio Goes, "Creative Culinary Recipe Generation Based on Statistical Language Models", in IEEE Access, 2020.
B. A. G. Oliveira, C. A. P. D. S. Martins, F. Magalhaes and Fabricio Goes, "Difference Based Metrics for Deep Reinforcement Learning Algorithms," in IEEE Access, vol. 7, pp. 159141-159149, 2019.
A. R. da Silva and Fabricio Goes, "HearthBot: An Autonomous Agent Based on Fuzzy ART Adaptive Neural Networks for the Digital Collectible Card Game HearthStone," in IEEE Transactions on Games (TOG), vol. 10, no. 2, pp. 170-181, 2018.
Fabricio Goes et al., "HoningStone: Building Creative Combos with Honing Theory for a Digital Card Game," in IEEE Transactions on Computational Intelligence and AI in Games (TCIAIG), vol. 9, no. 2, pp. 204-209, 2017.
A. Amorim, C. Franca, A. R. Silva, Fabricio Goes, "Creative Flavor Pairing: Design Flavor Combinations Using the Food Pairing Hypothesis and Assessing their Creativity by the RDC Metric", in the International Conference on Computational Creativity (ICCC), 2017.
H. A. Batista, C. Franca, Fabricio Goes, "Evolutionary Music Composition for Digital Games Using Regent-Dependent Creativity Metric", in the Brazilian Symposium on Games and Digital Entertainment (SBGames), 2017.
C. Franca, Fabricio Goes, A. Amorim and A. R. Silva, "Regent-Dependent Creativity: A Domain Independent Metric for the Assessment of Creative Artifacts", in the International Conference on Computational Creativity (ICCC), 2016.
Supervision
Teaching
Press and media
Due to my recent publications, I was recently invited to be interviewed about LLMs and Computational Creativity on the:
- Data Skeptic Podcast - the most popular data science podcast
- CBC Spark podcast - Canadian Broadcasting Corporation TV/Radio
- restofworld.org Tech News - a world emergent tech news